Artificial Intelligence In Detecting Normal-Tension Glaucoma

The aim of this study was to verify the AI–based model for early detection of normal tension glaucoma using the latest diagnostic tools and the advantages they provide, including the use of AI.Normal-tension glaucoma (NTG) is a subtype of open-angle glaucoma characterized by progressive optic neuropathy and corresponding visual field loss despite intraocular pressure (IOP) measurements within statistically normal ranges. Because elevated IOP is absent, NTG often remains underdiagnosed until significant structural and functional damage has occurred. Recent advances in artificial intelligence (AI), particularly deep learning and convolutional neural networks (CNNs), offer promising tools to enhance the early and accurate detection of NTG.

Conducted at a tertiary ophthalmology center using all the diagnostic methods and devices that we use during glaucoma diagnosis.

AI-based models have demonstrated high diagnostic performance using multimodal data, including color fundus photography, optical coherence tomography (OCT), and visual field testing. These systems can identify complex patterns of retinal nerve fiber layer thinning, ganglion cell complex alterations, and optic disc morphology that may not be readily apparent to clinicians. Integration of AI with imaging platforms such as Optical Coherence Tomography enhances automated structural assessment, while machine learning models applied to standard automated perimetry improve functional analysis.

Despite promising results, challenges remain, including the need for large, diverse datasets, external validation across populations, explainability of algorithms, and regulatory and ethical considerations. Additionally, clinical integration requires seamless workflow incorporation and clinician trust in AI-assisted decision-making.

In conclusion, AI represents a transformative approach in the detection of NTG, offering improved sensitivity for early disease identification and risk stratification. Continued research, multicenter validation, and responsible implementation are essential to fully realize its potential in clinical ophthalmology practice.